{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:QVBLCRQXF4LDBCFNQ5UFU4AQVJ","short_pith_number":"pith:QVBLCRQX","schema_version":"1.0","canonical_sha256":"8542b146172f163088ad87685a7010aa518b806bc70799cf1086e12bbc63264a","source":{"kind":"arxiv","id":"2208.01003","version":5},"attestation_state":"computed","paper":{"title":"What Can Be Learnt With Wide Convolutional Neural Networks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alessandro Favero, Francesco Cagnetta, Matthieu Wyart","submitted_at":"2022-08-01T17:19:32Z","abstract_excerpt":"Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harness the local and hierarchical structure of natural data such as images. Yet, we lack a quantitative understanding of how such structure affects performance, e.g., the rate of decay of the generalisation error with the number of training samples. In this paper, we study infinitely-wide deep CNNs in the kernel regime. First, we show that the spectrum of the corresponding kernel inherits the hierarchical structure of th"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2208.01003","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2022-08-01T17:19:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8ffed121fc53bad52881379b5a8f45bc3833dee9d7ffa5f5eaa6e6e534d6bf06","abstract_canon_sha256":"52cb518ba2d459fa41d15c0e6cb300ed9b47f6407fa150d3bf04787ad512bd9a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:16:11.184962Z","signature_b64":"pD6nvee02kIIkHU2mKzQCsbnt0MBibq3UZ9iQ798ypj8aoJy9BuQVT+342oOBfyEVM2/xScew+GBMTklMSDJCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8542b146172f163088ad87685a7010aa518b806bc70799cf1086e12bbc63264a","last_reissued_at":"2026-07-05T06:16:11.184524Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:16:11.184524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Can Be Learnt With Wide Convolutional Neural Networks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alessandro Favero, Francesco Cagnetta, Matthieu Wyart","submitted_at":"2022-08-01T17:19:32Z","abstract_excerpt":"Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harness the local and hierarchical structure of natural data such as images. Yet, we lack a quantitative understanding of how such structure affects performance, e.g., the rate of decay of the generalisation error with the number of training samples. In this paper, we study infinitely-wide deep CNNs in the kernel regime. First, we show that the spectrum of the corresponding kernel inherits the hierarchical structure of th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.01003","kind":"arxiv","version":5},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2208.01003/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2208.01003","created_at":"2026-07-05T06:16:11.184581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.01003v5","created_at":"2026-07-05T06:16:11.184581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.01003","created_at":"2026-07-05T06:16:11.184581+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVBLCRQXF4LD","created_at":"2026-07-05T06:16:11.184581+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVBLCRQXF4LDBCFN","created_at":"2026-07-05T06:16:11.184581+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVBLCRQX","created_at":"2026-07-05T06:16:11.184581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.09350","citing_title":"The Cost of Discretization in Functional Linear Regression: Minimax Rates and Adaptation","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ","json":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ.json","graph_json":"https://pith.science/api/pith-number/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/graph.json","events_json":"https://pith.science/api/pith-number/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/events.json","paper":"https://pith.science/paper/QVBLCRQX"},"agent_actions":{"view_html":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ","download_json":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ.json","view_paper":"https://pith.science/paper/QVBLCRQX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.01003&json=true","fetch_graph":"https://pith.science/api/pith-number/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/graph.json","fetch_events":"https://pith.science/api/pith-number/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/action/storage_attestation","attest_author":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/action/author_attestation","sign_citation":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/action/citation_signature","submit_replication":"https://pith.science/pith/QVBLCRQXF4LDBCFNQ5UFU4AQVJ/action/replication_record"}},"created_at":"2026-07-05T06:16:11.184581+00:00","updated_at":"2026-07-05T06:16:11.184581+00:00"}